Text Classification
Transformers
TensorBoard
Safetensors
English
distilbert
Generated from Trainer
Eval Results (legacy)
text-embeddings-inference
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("Hartunka/tiny_bert_km_20_v2_qnli")
model = AutoModelForSequenceClassification.from_pretrained("Hartunka/tiny_bert_km_20_v2_qnli")Quick Links
tiny_bert_km_20_v2_qnli
This model is a fine-tuned version of Hartunka/tiny_bert_km_20_v2 on the GLUE QNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.6494
- Accuracy: 0.6279
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 256
- eval_batch_size: 256
- seed: 10
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6674 | 1.0 | 410 | 0.6506 | 0.6101 |
| 0.6407 | 2.0 | 820 | 0.6511 | 0.6229 |
| 0.5998 | 3.0 | 1230 | 0.6494 | 0.6279 |
| 0.5319 | 4.0 | 1640 | 0.7139 | 0.6154 |
| 0.4555 | 5.0 | 2050 | 0.7635 | 0.6118 |
| 0.3843 | 6.0 | 2460 | 0.9000 | 0.6072 |
| 0.3163 | 7.0 | 2870 | 1.0751 | 0.6050 |
| 0.2599 | 8.0 | 3280 | 1.1298 | 0.6077 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1
- Downloads last month
- 4
Model tree for Hartunka/tiny_bert_km_20_v2_qnli
Base model
Hartunka/tiny_bert_km_20_v2Dataset used to train Hartunka/tiny_bert_km_20_v2_qnli
Evaluation results
- Accuracy on GLUE QNLIself-reported0.628
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Hartunka/tiny_bert_km_20_v2_qnli")